Professional Documents
Culture Documents
Diedr Mller
Johannesburg
July 2007
ABSTRACT
Customer satisfaction and retention are critical for retail banks. Service quality
is a major determinant of customer satisfaction, and is increasingly being seen
as a key strategic differentiator within the financial services sector worldwide.
However, little is known about service quality in an African context for retail
banking.
Finally, the research demonstrates that service expectations in Africa are not
static. Service expectations do change over time, and this change is evident
over a relatively short period of time (< 1 year).
Diedr Mller
30 July 2007
DEDICATION
ACKNOWLEDGEMENTS
Alana Moller, Anne Du Plessis, Michael Franze and Carl Grim, for your
incredible wisdom, leadership and unremitting love.
Dr. Geoff Bick, and Dr. Anthony Stacey of the University of the Witwatersrand,
Graduate School of Business Administration.
This research would not have been possible without their support, guidance and
mentorship.
TABLE OF CONTENTS
1 INTRODUCTION.........................................................................1
1.1 PURPOSE OF STUDY ........................................................................................ 1
1.2 CONTEXT OF STUDY ........................................................................................ 2
1.3 PROBLEM STATEMENT ..................................................................................... 2
1.3.1 SUB-PROBLEM ONE:.............................................................................................. 2
1.3.2 SUB-PROBLEM TWO: ............................................................................................. 3
1.3.3 SUB-PROBLEM THREE: .......................................................................................... 3
1.4 SIGNIFICANCE OF STUDY.................................................................................. 3
1.5 DELIMITATIONS AND LIMITATIONS...................................................................... 3
1.5.1 DELIMITATIONS ..................................................................................................... 3
1.5.2 LIMITATIONS ......................................................................................................... 4
i
4 PRESENTATION, ANALYSIS AND INTERPRETATION OF
RESULTS...........................................................................................41
4.1 DEMOGRAPHIC DATA ..................................................................................... 41
4.2 CROSS NATIONAL DIFFERENCES .................................................................... 43
4.3 IMPORTANCE RANKING OF SERVICE DIMENSIONS AND ATTRIBUTES ................... 46
4.3.1 SIGNIFICANCE OF ATTRIBUTE MEANS .................................................................... 53
4.4 CHANGES IN SERVICE EXPECTATIONS OVER TIME (Q1 AND Q4 2006) .............. 62
REFERENCES...................................................................................75
ii
LIST OF TABLES
TABLE 1: SERVQUAL DIMENSIONS .......................................................................... 19
TABLE 2: DEFINITIONS OF SERVQUAL DIMENSIONS ................................................. 20
TABLE 3: CLASSIFICATION OF ATTRIBUTES FOR RETAIL BANKS.................................... 22
TABLE 4: RANKING OF SERQUAL DIMENSIONS BY CUSTOMERS IN THE RETAIL BANKING
SECTOR ........................................................................................................ 24
TABLE 5: MAPPING OF SERVICE ATTRIBUTES TO SERVQUAL DIMENSIONS ................ 29
TABLE 6: SURVEYED COUNTRIES AND RESPECTIVE RETAIL BANKS.............................. 32
TABLE 7: SAMPLE SIZE BY COUNTRY ......................................................................... 33
TABLE 8: QUESTIONNAIRE TRANSLATION BY COUNTRY ............................................... 35
TABLE 9: RESULTS OF ONE-WAY ANOVA, Q4 2006 ................................................. 45
TABLE 10: IMPORTANCE RANKING OF OVERALL MEANS, BY SERVICE DIMENSIONS AND
ATTRIBUTES, Q4 2006 .................................................................................. 47
TABLE 11: IMPORTANCE RANKING OF STANDARDISED MEANS, BY COUNTRY .................. 52
TABLE 12: SIGNIFICANCE OF ATTRIBUTE MEANS BY COUNTRY, Q4 2006..................... 55
TABLE 13: T-TEST FOR SIGNIFICANCE OF DIFFERENCES BETWEEN OVERALL SERVICE
ATTRIBUTE MEANS OVER TIME (Q1 - Q4 2006) ................................................ 63
TABLE 14: T-TEST FOR SIGNIFICANCE OF DIFFERENCES BETWEEN COUNTRY ATTRIBUTE
MEANS OVER TIME (Q1 - Q4 2006) ................................................................. 64
TABLE 15: IMPORTANCE RANKING OF SERQUAL DIMENSIONS ................................... 67
TABLE 16: IMPORTANCE RANKING OF SERVICE ATTRIBUTES ....................................... 69
TABLE 17: IMPORTANCE RANKING OF SERVICE ATTRIBUTES Q1 2006 VS Q4 2006... 71
TABLE 18: SIGNIFICANT CHANGES IN MEAN IMPORTANCE BETWEEN Q1 2006 Q4 2006
.................................................................................................................... 72
iii
LIST OF FIGURES
FIGURE 1: THE SERVICE-PROFIT CHAIN....................................................................... 8
FIGURE 2: EXPECTATIONS MANAGEMENT MODEL ....................................................... 15
FIGURE 3: SERVICE GAP MODEL OF THE SERVICE PROCESS ...................................... 17
FIGURE 4: CLASSIFICATION OF SERVQUAL SERVICE DIMENSIONS ............................ 21
FIGURE 5: DATA ANALYSIS SPIRAL ............................................................................ 38
FIGURE 6: RESPONDENT REPRESENTATION BY COUNTRY, Q1 2006 ........................... 42
FIGURE 7: RESPONDENT REPRESENTATION BY COUNTRY, Q4 2006 ........................... 42
FIGURE 8: RESPONDENT REPRESENTATION BY BANK, Q4 2006.................................. 43
FIGURE 9: IMPORTANCE RANKING OF POSITIVE OVERALL MEANS BY SERVICE
DIMENSION, Q4 2006 .................................................................................... 48
FIGURE 10: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR BOTSWANA ...... 57
FIGURE 11: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR EGYPT ............. 57
FIGURE 12: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR GHANA ............ 58
FIGURE 13: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR KENYA ............. 58
FIGURE 14: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR MAURITIUS ....... 59
FIGURE 15: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR SEYCHELLES .... 59
FIGURE 16: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR TANZANIA......... 60
FIGURE 17: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR UGANDA .......... 60
FIGURE 18: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR ZAMBIA ............ 61
FIGURE 19: SIGNIFICANT SERVICE DIMENSIONS AND ATTRIBUTES FOR ZIMBABWE ....... 61
i
1 INTRODUCTION
There is little doubt that achieving customer satisfaction and loyalty is essential for
long-term survival. However, to achieve customer satisfaction, a superior level of
service and customer orientation is required (Bick, Brown & Abratt, 2004).
It therefore stands to reason that in order for retail banks to thrive, both product
and service delivery must be adequately aligned with customer expectations.
-1-
2. How stable are the service expectations? Are they constant or do they
change over time?
According to Reichheld & Sasser (1990) the cost of gaining a new customer is
about five times greater than the cost of retaining a current customer through the
use of relationship marketing. Newman & Cowling (1996: 106) noted that, "for UK
financial institutions, it is estimated that an increase of 5 per cent in customer
retention is potentially worth 100 million a year. Realisations such as these have
attracted in the mid-1990s the attention of directors of retail banks to measuring
their service quality and customer satisfaction and initiating major service quality
change programmes".
-2-
1.3.2 Sub-problem Two:
The research also provides guidance for commercial bank managers wanting to
understand the dynamics of customer service expectations in Africa, and improve
the quality of service delivered, in order to enhance customer satisfaction and
customer loyalty.
1.5.1 Delimitations
The scope of this study is limited to customers from the top five/ six commercial
banks within the following ten countries:
Botswana
Egypt
-3-
Ghana
Kenya
Mauritius
Seychelles
Tanzania
Uganda
Zambia
Zimbabwe
1.5.2 Limitations
The study is limited to retail banks, and excludes corporate and merchant banks.
Research was only conducted in principal cities of each country.
The research only explores service expectations of retail banking customers, and
does not explore the service gaps that exist between customer expectations and
customer experience (perceived services).
-4-
2 LITERATURE REVIEW
2.1 Introduction
The African financial services sector has recently seen a growing intensity of
competition within the marketplace. No organisation can afford to overlook the
importance of customer satisfaction, which is essential for developing customer
loyalty and subsequent profitability within the banking sector.
-5-
mouth directly affect the viability and profitability of a firm (Dabholkar, Thorpe &
Rentz, 1996).
Customer satisfaction, and related retention rates, can come from a range of
activities available to the firm. Levesque et al (1996) points out that for retail
banks, major gains in customer satisfaction are likely to come from improvements
in: (i) service quality; (ii) service features; and (iii) customer complaint handling.
For this investigation, service quality is used as the key measure towards
improving customer satisfaction.
-6-
Service quality is widely recognised as being a critical determinant for the success
of an organisation in todays competitive environment. Any decline in customer
satisfaction due to poor service quality would be a matter of concern. Consumers
being more aware of rising standards in service, prompted by competitive trends,
have developed higher expectations (Marshall, Baker & Finn, 1998: 381).
As a result, there has been a growing interest in service quality issues, which are
both internal and external to an organisation. For example, Heskett, Jones,
Loveman, Sasser, & Schlesinger (1994) proposed a service-profit chain (SPC)
that integrates these perspectives into a model that establishes relationships
between internal service quality and employee satisfaction, external service
quality and customer satisfaction, and profitability.
The links in the SPC are as follows: profit and growth are stimulated primarily by
customer loyalty. Loyalty is a direct result of customer satisfaction. Satisfaction is
largely influenced by the value of services provided to customers. Value is created
by satisfied, loyal, and productive employees. Employee satisfaction, in turn,
results primarily from high-quality support services and policies that enable
employees to deliver results to customers. (See Figure 1).
-7-
Figure 1: The Service-Profit Chain
Revenue
Growth
External Customer Customer
Service Satisfaction Loyalty
Value Profitability
Service designed Retention
Service and delivered to Repeat
concept; meet targeted Business
results for customers needs Referral
customers
The popularity of the SPC framework is evident in the numerous case studies
reported by practitioners, the most notable being its application at Sears (Rucci,
Kirn & Quinn, 1998).
-8-
Rust, Zahorik & Keiningham (1995) proposed another framework; "Return on
Quality". This model is similar, although it differentiates itself by explicitly
modelling the cost and benefits of quality related investments. In the SPC
framework, the focus is on revenue rather than profitability, while the ROQ model
focuses on the cost (and return) of quality.
However, both the SPC and ROQ have several commonalities, the most notable
being an emphasis on driving the firm's operations based on statistical analyses of
customer surveys. Thus, customer surveys are used to identify key service
attributes (and expectations) that impact overall quality and retention, which in
turn drive financial outcomes. In this respect, they can provide actionable
guidance to management (Kamakura, Mittal, De Rosa & Mazzon, 2002).
Quality is nowadays among the most critical aspects for the strategic
management of service firms (Robledo, 2001). Perceived quality reflects the
opinion of the customer regarding the superiority or global excellence of a product
or service (Zeithaml, Berry & Parasuraman, 1988)
According to Venetis & Ghauri (2004), service quality is regarded as one of the
few means of services differentiation, attracting new customers and increasing
market share. It is also viewed as an important means of customer retention. The
lifetime value of a loyal customer can be astronomical, especially when referrals
-9-
are added to the economics of customer retention and repeat purchases of related
products (Heskett et al, 1994)
In one case, a retail bank that increased its customer retention rates by 5 per cent
increased its profits by 85 per cent (Reichheld & Sasser, 1990). The cost of
gaining a new customer is about five times greater than the cost of retaining a
current customer through the use of relationship marketing. Newman & Cowling
(1996: 106) noted that, "for UK financial institutions, it is estimated that an
increase of 5 per cent in customer retention is potentially worth 100 million a
year.
The ability to provide a quality service will, therefore, improve a commercial banks
ability to increase market share and profitability, whilst at the same time reducing
their existing customers switching propensity to another institution. Thus the
ability to consistently provide a high quality service offering may well act as a key
strategic differentiator.
- 10 -
perceived as the highest expectation based on research undertaken in the UK,
whilst "reliability" was the most important quality dimension to US consumers.
Smith & Reynolds (2001: 405) state that "Cultural differences may also explain
variations in consumers' reported perceptions of a service".
- 11 -
Speece & Pinkaeo (2002) described the differences in customer expectations
based on ethnocentrism in Thailand compared to other countries. Espinoza (1999)
indicates that service managers need to be cognizant of the parts of the service
delivery experience that are open to cultural influences as contrasted with those
that remain stable across cultures. For example, in his study Assessing the
cross-cultural applicability of a service quality measure, he finds responsiveness
was the most important dimension for Quebecers while tangibles was the most
important for Peruvians. Differences in service expectations are relevant and need
to be considered when designing a marketing strategy.
2.5.1 Conclusion
In summary, the literature suggests that cross-national differences exist and are
key to designing a marketing strategy, however little research has been done in
Africa on this topic.
- 12 -
2.6 Service Quality and Service Expectations
- 13 -
customer expectations may be too high relative to performance, or vice versa).
Doing this requires an understanding of the sources of customer expectations.
Sources of expectations:
- 14 -
Figure 2: Expectations Management Model
- 15 -
2.6.2 Measuring service quality and customer expectations
Much literature exists on quality measurement. This section discusses the models
used to measure service quality and the classification of quality dimensions.
The most common models used for measuring service quality are SERVQUAL,
developed by Parasuraman, Zeithaml & Berry (1985, 1988, 1990, 1993), and
SERVPERF, developed by and Cronin & Taylor (1992).
- 16 -
Figure 3: Service Gap Model of the Service Process
Expected Service
Gap 1
Management perception of
expected service
Gap 2
Gap 5
Service quality
specification
Gap 3
Gap 4
Service Delivery Communication to
customer
Perceived service
- 17 -
Quester & Romaniuk (1997) found that little difference existed between the
SERVQUAL and SERVPERF models, based on a comparative study conducted in
the Australian advertising industry.
Grnroos (1988) identifies the five key determinants of service quality as:
- 18 -
accessibility and flexibility; and
The first of these are technical (outcome related) and the last four are functional
(process related).
reliability
tangibles
responsiveness
assurance
empathy
SERVQUAL Components
Tangibles Tangibles
Reliability Reliability
Responsiveness Responsiveness
Competence
Courtesy
Assurance
Credibility
Security
Access
Empathy Communication
Understanding
- 19 -
Table 2: Definitions of SERVQUAL Dimensions
SERVQUAL Definition
Dimensions of
Service
Blanchard & Galloway (1994) describe an alternative basis for modelling service
quality. As a result of the possible overlap and ambiguity in the SERVQUAL
dimensions, Blanchard et al (1994) proposed a model based on the three
dimensions of process/outcome, subjective/objective, and soft/hard. These are, in
the authors opinion probably orthogonal. Subjective/objective provides a
measure of the degree to which the quality of an aspect of service can be
objectively specified. For example, cash availability in machines is objective,
whereas politeness of staff is more subjective. The terminology Soft/hard
describes hard as physical aspects of the service while soft represents
interpersonal interaction. Typically, soft issues (e.g. staff attitude) are more difficult
to manage and measure (Vandermerwe, 1993).
- 20 -
Blanchard & Galloway (1994) demonstrate that process is far more important
than outcome in determining customer perceptions of service quality. They also
demonstrates that, objective and hard aspects in retail banking services are as
important as the subjective and soft aspects, but can be much more readily
identified and specified. Thus addressing the hard and objective issues could
more readily provide, and monitor, a high quality service in a fairly prescriptive and
reproducible manner (Blanchard et al, 1994: 14).
Expectation
Outcome
Corporate image
Tangibles Reputation and credibility
Premises Assurance
Tangibles
Reliability
Interactive behaviour
and attitudes
Responsiveness
Empathy
Process
- 21 -
Table 3: Classification of Attributes for Retail Banks
Privacy of discussions P 0 H
Politeness of counter staff P SIO S
Willingness to help P SIO S
Way staff treat customers P S S
Cash in machines 0 0 H
Speed and efficiency of transactions P/O SIO SIH
Staff listen P S S
Tills open at busy times P 0 H
Staff available to help P SIO H
Informing of account changes P/O 0 H
Way mistakes are handled P/O S S
Clarity of staff answers P/O S S
Charges clear and explained 0 SIO SIH
Attitude of bank when lending P S S
Person for questions P 0 H
Availability of enquiries desk P 0 H
Return/replace cash card 0 0 H
Queue at branch P SIO H
Opening hours P 0 H
Appearance of branch P SIO H
Taking time to match product/need P SIO SIH
Staff's product knowledge P 0 SIH
Overdraft charges 0 0 H
Ease of getting to branch P SIO H
Queues at cash machines P SIO H
Ease of understanding letters P SIO SIH
Informed of progress of applications P/O SIO H
Relevant direct mail 0 SIO H
Telephones answered quickly P 0 H
Getting right person on phone P 0 SIH
Speed of mortgage confirmation P SIO H
Parasuraman et al (1990) points out that while reliability is largely concerned with
the service outcome, tangibles, responsiveness, assurance and empathy are
more concerned with the service process. He reported in his quantitative
- 22 -
research that regardless of the service being studied, customers have always
favoured reliability when asked to indicate the relative importance of the five
dimensions in judging service.
Zeithaml et al. (1990) also maintained that the highest expectation is reliability.
Reliability of service comes in first regardless of how we measure salience, or
which specific service we study. This has led them to the conclusion that because
reliability scores the highest, it must be the most important.
However, Johnson & Mathews (1997) found that Security comes out as the
highest service quality expectation. Other research undertaken in the UK has also
identified this discrepancy. For example, Mathews (1995) finds that reliability does
not come out as the most important expectation score, while Berry et al. (1985)
and Johnston (1995) identified responsiveness as an important factor.
The Parasuraman et al. (1990) study asked users of credit cards and retail
banking services, amongst users of other products/ services, to rate the
importance of each service dimensions. The results indicated that reliability of
service was the most important requirement of these financial service customers.
The second most important requirement for credit card customers was
responsiveness and bank customers empathy.
- 23 -
Table 4: Ranking of SERQUAL Dimensions by Customers in the Retail
Banking Sector
2.6.4 Conclusion
In summary, the literature suggests that service quality is derived from service
expectations, and highlights the importance of service quality as a strategic
differentiator in facilitating customer satisfaction and retention. Responsiveness
has been shown to be an important factor, supported by previous work from Berry
et al. (1985), Johnston (1995) and Bitner et al. (1990). Furthermore, Avkiran
(1994), Blanchard et al (1994), and Johnston (1997) recognise responsiveness as
the most important service dimension in the retail banking sector.
1. Responsiveness
2. Assurance
3. Empathy
4. Tangibles
5. Reliability
- 24 -
2.7 Changes in Service Expectations over Time
Quality of service is not an absolute concept. Quality of service changes with time,
experience and attitude, and can be linked to customer value (Woodruff, 1997).
2.7.1 Conclusion
2.8.1 Proposition 1
This proposition is based on the research by Johnson & Mathews (1997), Webster
(1989), Davidow & Uttal (1989), Knight (1999), Stauss & Mang (1999), Smith &
- 25 -
Reynolds (2001), Lovelock (1999), and Speece & Pinkaeo (2002) who
acknowledge cross-national cultural differences.
2.8.2 Proposition 2
1. Responsiveness
2. Assurance
3. Empathy
4. Tangibles
5. Reliability
Parasuraman et al. (1985, 1988, 1990, 1993, 1994), indicated that quality of
service delivered can be classified in five major dimensions, namely; tangibles,
reliability, responsiveness, assurance and empathy.
2.8.3 Proposition 3
- 26 -
3 RESEARCH METHODOLOGY
This chapter describes the methodology that will be used to answer the research
propositions posed in the previous section. The chapter starts off with a
description of the chosen method and its implications for this research. The
research design, population description and sample selection are then discussed.
This is followed by descriptions of the processes used for data collection, analysis
and interpretation. The chapter is concluded with a discussion on validity and
reliability of the chosen research method.
Quantitative research using nomothetic commitments was best suited for this
research as the study follows on from previous empirical data (SERVQUAL,
Blanchard & Galloway (1994), which identifies what the key service attributes are
in retail banking. Hence a qualitative interpretive approach in order to capture the
subjects perspectives is not required. Rather, the objective of the research
methodology was to ensure that the outcomes have a greater degree of reliability
and objectivity, and are less impressionistic, which can be achieved through the
usage of quantitative research (Denzin & Lincoln, 1998).
A comparative scaling technique applying a partial rank order scale was used in
the questionnaire to determine the top three most important service attributes (in
order of importance), out of fifteen attributes.
- 27 -
The use of rank ordering of survey items in the questionnaire is suitable since it is
both intuitive for respondents and is independent of any particular response scale.
However, a disadvantage of rank order scales is the complexity of rigorous
analysis (Stacey, 2006). In order to appropriately apply parametric statistics
(which assumes interval data), the ordinal data derived from this scaling technique
was rescaled using the algorithmic approach to analysing rank ordered survey
items (Stacey, 2006). This allowed for a more accurate estimation of the
population item means and standard deviations.
The data was collected in Q1 2006 and Q4 2006 by Synovate (leading market
research consultants). The initial study was conducted under the supervision of
the researcher.
Banks, in seeking to provide a high quality service, should identify needs and
expectations and establish the way in which customers prioritise them (Blanchard
& Galloway, 1994). Parasuraman et al. (1990) proposes a model of the
determinants of service quality. The model provides a framework for analysing
quality failures by identifying the service gaps that arise, which cause a
mismatch between customer expectation and customer experience (perceived
service), shown in Figure 3. This study focuses purely on the understanding and
prioritisation of customer service expectations, which forms the fundamental base
level of the model.
- 28 -
The questionnaire was adapted from the SERVQUAL dimensions i.e. reliability,
tangibles, responsiveness, assurance, and empathy. Table 5 below shows the
mapping of the fifteen service attributes to the SERVQUAL dimensions.
communication material
Provides a welcoming environment
- 29 -
The caring individualised
Focus on building long term
attention provided to the
relationships
customer.
Empathy
- Understanding
Understands my personal banking
- Communication
needs
- Access
The first part of the questionnaire addressed demographic data which will be used
for narrative purposes only.
The second part of the questionnaire used a partial rank order scale, which lists
fifteen key service attributes in accordance with the SERVQUAL requirements
(Parasuraman et al, 1994). Respondents were asked to rank the top three
attributes in order of importance.
A partial rank order scale is suitable for hypothesis testing once rescaled using the
algorithmic approach presented by Stacey (2006). Furthermore, it was selected in
order to avoid respondent fatigue. The task of rank ordering large numbers of
items can be unduly onerous for respondents, which has a negative impact on the
validity and reliability of the data (Stacey, 2006).
Hypothesis testing using analysis of variance (ANOVA) was used to compare the
means of ten different populations; where :
Ho: 1 = 2 = 3 = 4 = 5 = 6 = 7 = 8 = 9 = 10
Ha: At least one pair of means are not equal,
Assumptions:
- 30 -
The samples must be selected from normally distributed populations
3.2.1 Population
The targeted population for this study was limited to retail banking customers,
from the capital cities of the following ten countries and respective banks, listed
overleaf:
- 31 -
Table 6: Surveyed Countries and Respective Retail Banks
Egypt Barclays
Bank MISR NSGB Citigroup HSBC CIB
- 32 -
3.2.2 Sample
Seychelles 75 200
Sample Selection
The use of some basic selection rules guided the intercepts. The interviews were
done mainly during the week. The day was then divided into time slots and a time
slot was randomly selected to start interviewing in. A person was intercepted
every nth minute for an interview to ensure randomness.
- 33 -
Sampling Substitution:
Sampling Contingency:
Some over-sampling was done for the surveys in order to meet the required
samples for each country. This catered for questionnaires which are incomplete.
This was managed with local suppliers to ensure standard errors (at 90% level of
confidence) at regional level.
Sampling Error
The phenomenon of sampling error is that there is unavoidable and random error
or difference between the true (unknown) population parameters and those
parameters which have been estimated or inferred from the sample (Stacey,
2006). In this case, the larger sample sizes ensured a better representation of the
population as a whole and hence reduce the potential for sampling error.
Since customer databases were not available for the envisaged samples, it was
necessary to use face-to-face interviews as a means of contact. This technique
allowed for a higher response rate and is more convenient for the respondent.
However, apart from the economical and logistical disadvantages of this
technique, it should be kept in mind that some respondents would still give biased
responses when face-to-face with a researcher. Furthermore, language translation
may reduce face validity.
- 34 -
interview. A short structured questionnaire not longer than 15 minutes was used.
Please refer to Appendix 1 which contains an outline of the questionnaire.
Test data files were sent to South Africa after the first few interviews for
centralised checking of data input
Country Translation
Botswana English / Setswana
Egypt Arabic
Ghana English / Twi
Kenya English / Kiswahili
Mauritius English / French and Mauritian Creole
Seychelles English / Seychelles Creole
Tanzania English / Kiswahili
Uganda English / No translation
Zambia English / No translation
Zimbabwe English / No translation
- 35 -
Research was conducted in accordance with the code of ethics prescribed
by the European Society for Opinion and Marketing Research (ESOMAR),
and the South African Marketing research Association (SAMRA).
Respondents were required to rank order (in order of importance) the top three
most important service attributes relating to their bank, out of 15 service attributes
in total. The data was summarised by tabulating the frequency (and proportion) of
occurrence of each preference permutation.
Rank ordered data are only ordinal in nature because respondents give no
indication of the magnitude of the difference between consecutively ranked items.
However the use of parametric techniques using at least interval data are required
to compare mean rankings of the partially rank ordered survey items.
Once the mean and standard deviations are derived, the proposed methodology
makes use of hypothesis ANOVA testing, t-tests and importance ranking to
determine whether cross-national differences in service expectations exist, and
what service attributes proposed by retail banks are considered most important to
customers. Hypothesis t-testing is used to determine whether service expectations
are constant over time.
- 36 -
In summary, data analysis was conducted as follows:
- The ordinal data derived from this scaling technique (partial rank order)
was rescaled using the algorithmic approach to analysing partial rank
ordered survey items (Stacey, 2006). This allowed for the estimation of the
population item means and standard deviations in order to perform
hypothesis testing
- Two sample hypothesis t-tests were carried out for each attribute in order
to determine the significance of differences between country means over
time
The above steps were integrated with Creswells data analysis spiral as described
in Leedy & Ormrod (2001:161).
- 37 -
Figure 5: Data Analysis Spiral
Validity is concerned with whether the instrument (in this case surveys) measures
what it is supposed to measure (Leedy & Ormrod, 2001: 31) and whether it will
lead to valid conclusions about cross-national service expectations in retail
banking. Since quantitative survey research was conducted, the following issues
around validity are evident. These issues are addressed below in the discussion
on the various types of validity.
External validity refers to generalisability, i.e. the ability for the research to be
generalised across persons, settings and times.
- 38 -
Large sample sizes were used in conjunction with a random sampling
methodology which is probability based (Leedy & Ormrod, 2001), thus it is
possible to generalise the results of the study.
The author did not have direct control over the collection of the data, which may
be a limitation in the research. However quality controls were put in place (as
discussed above) in order to ensure validity.
Internal validity refers to the extent to which the instrument allows inferences
about the causal relationships between data elements (Leedy & Ormrod, 2001).
There are various types of internal validity, namely: content validity, construct
validity and criterion-related validity (Leedy & Ormrod, 2001). Content and
construct validity is relevant and therefore discussed here.
Content validity was addressed by ensuring that there were adequate questions to
completely cover all the relevant aspects identified in the literature review.
The task of rank ordering large numbers of items can be unduly onerous for
respondents, which has a negative impact on the validity and reliability of the data
(Stacey, 2006). This was addressed by using partial rank ordering in the
questionnaire in order to reduce respondent fatigue.
- 39 -
3.5.3 Reliability
The quality controls (discussed above) also ensured that the research is
administered consistently throughout all the countries.
A possible limitation was that the sample chosen was more heavily weighted
towards Barclays customers vs. the other banks customers, since the study was
commissioned by Barclays PLC. In addition, the study was only conducted in the
capital cities of each country.
- 40 -
4 PRESENTATION, ANALYSIS AND INTERPRETATION OF
RESULTS
The results of the analysis are presented in this chapter. A more detailed
breakdown of the results is presented in Appendix B.
Sample sizes for all countries were large enough to ensure adequate
representation of the total population, thus reducing the potential for sampling
error.
- 41 -
Figure 6: Respondent Representation by Country, Q1 2006
Zimbabwe Botswana
12% 15%
Zambia Egypt
12% 7%
Uganda Ghana
7% 12%
Tanzania
Mauritius Kenya
7%
10% 14%
Seychelles
4%
Total Sample: 2075
Zimbabwe Botswana
12% 13%
Zambia Egypt
12% 8%
Uganda Ghana
8% 12%
Tanzania Kenya
6% Mauritius
15%
9%
Seychelles
5%
- 42 -
Figure 8: Respondent Representation by Bank, Q4 2006
Other Banks
26%
Barclays
74%
Users of retail banking services were read a list of 15 service attributes relating to
their bank, and asked to rank the top three most important service attributes. The
data was summarised by tabulating the frequency (and proportion) of occurrence
of each preference permutation, as illustrated in Appendix C.
In order to apply parametric testing, the ordinal data derived from the partial rank
order was rescaled using the algorithmic approach to analysing partial rank
ordered survey items (Stacey, 2006). The shotgun stochastic parameter
estimation algorithm iteratively and asymptotically estimates the population means
and standard deviations of the 15 survey items from which the observed data
sample was drawn.
This allowed for the estimation of underlying parameters (population item means
and standard deviations) of partially ranked survey items in order to perform
hypothesis testing.
- 43 -
ANOVA testing on the rescaled standardised means and standard deviations was
used to determine if there are differences across the ten African countries in the
perceived importance for each attribute.
The results are based on the assumptions that each population is approximately
normally distributed and there are equal population variances. Central Limit
theorem can be used as the basis for normality due to large sample sizes. The
samples were independent and selected at random. A significance level of 5%
was used.
The findings in Table 9 below show that in all cases (15 attributes), the zero
hypothesis is reject based on the low p values. Thus, there is sufficient evidence
to indicate that for every service attribute, one or more of the population (country)
means are not equal to the others. Every attribute had some significant difference
between countries. This implies that different cultural groups give different
importance to service quality dimensions.
This finding supports the proposition that customer service expectations in retail
banking differ significantly between countries in Africa. It is therefore critical to
take cross-national differences into consideration when designing and
implementing a marketing strategy for multinational companies. In order to
achieve this, country-specific customer expectations must be investigated and
prioritised as a first step towards improving service quality.
- 44 -
Table 9: Results of One-Way ANOVA, Q4 2006
Note assumptions:
The data has been standardised, therefore the overall mean = 0, and the overall standard deviation = 1
Each of the samples are drawn from a normal population
Central Limit theorem can be used as the basis for normality due to large sample sizes
The samples are independent and selected at random
Variance (or standard deviation) of the items (populations) are equal
Significance level: 5% ( = 0.05)
- 45 -
4.3 Importance Ranking of Service dimensions and Attributes
In order to better understand what form these cross-national differences take, the
following tests were carried out in order to highlight the similarities and differences in
customer service expectations.
Respondents ranked in descending order (i.e. first most important, second most
important and third most important) the top three most important service attributes
relating to their bank, out of 15 service attributes in total. The data was summarised by
tabulating the frequency (and proportion) of occurrence of each preference permutation,
as illustrated in Appendix C.
These results were converted into standardised sample means using the Stochastic
Search Algorithm (Stacey, 2006). The data was then ranked by service attribute and the
underlying service dimension. In the ranking, the higher the mean, the more important
that aspect of service delivery is to the respondents. Negative means have a less than
average importance. An overall mean was derived from the individual country means. No
weighting was applied to the overall mean in order to give each country equal
importance.
The individual attributes have been clustered into the underlying SERVQUAL
dimensions, and the overall mean for all countries is presented in Table 10 below. Figure
9 represents positive overall means and respective service dimensions graphically.
- 46 -
Table 10: Importance Ranking of Overall Means, by Service Dimensions and
Attributes, Q4 2006
Note:
The overall mean was derived from the individual country means
For the overall mean, no weighting was applied in order to give each country equal importance
The data has been standardised, therefore the overall mean = 0, and the overall standard
deviation = 1
The higher the mean, the more important that aspect of service delivery is to the respondents
Negative means are inconsequential since they represent a less than average importance
- 47 -
Figure 9: Importance Ranking of Positive Overall Means by Service Dimension, Q4
2006
0.800
0.700
0.600
0.500
0.400
0.300
0.200
0.100
0.000
Understands personal
Response time to info
Efficient staff
Friendly staff
Shorter queues
Convenient locations
Reliability
banking needs
requests
Note:
Negative means have been excluded since they represent a less than average importance
The overall mean was derived from the individual country means
No weighting was applied to the overall mean in order to give each country equal importance
- 48 -
Overall results (for Africa)
Results from the analysis of the SERVQUAL dimensions clearly indicate that
responsiveness is the most important service requirement for retail banking customers,
followed by reliability of service, tangibles, assurance and empathy. This is based
on the visual interpretation from Figure 9, where it is evident that most of the items
relating to responsiveness have the largest values. We can therefore assume that
responsiveness has the most important rating.
Analysis of the individual attributes indicate that staff efficiency is by far the most
important service criteria for retail banking customers in Africa. Shorter queues, service
reliability and convenient locations are the second, third and fourth most important
attributes respectively.
Dimension Responsiveness
Three out of five responsiveness attributes are ranked in the top 5. It is interesting to
note that respondents assign considerably more importance to staff efficiency than any
other attribute. On average, respondents perceive staff efficiency to be twice as
important in comparison to the second highest ranked attribute (shorter queues).
While responsiveness is typically process based, one could argue that in this case both
attributes driving this dimension relate to the time aspect of service delivery (rather than
willingness to help). Thus, this dimension is more outcomes based, and could be
classified as an objective hard issue.
- 49 -
Dimension Reliability
The underlying service dimension reliability reflects issues relating to the banks ability
to perform the promised service dependably and accurately.
While the literature review showed that reliability as a dimension is typically ranked fifth
in a retail banking environment, the research results show that in an African
environment, this dimension is ranked second most important by users of retail banking
services.
Reliability is outcome based (Gronroos, 1998) and can be classified as an objective hard
issue since it can be much more readily identified and specified to that of soft and
subjective relational issues. This implies that addressing the hard objective issues
could more readily provide, and monitor, a high quality service in a fairly prescriptive and
reproducible manner. (Blanchard & Galloway, 1994)
Dimension Tangibles
The underlying service dimension tangibles reflects issues relating to the appearance
of physical facilities, location and personnel.
The tangibles dimension is ranked third, whilst the literature review suggests this
dimension is typically ranked fourth.
Dimension Assurance
The underlying service dimension assurance reflects issues relating to the knowledge
and courtesy of employees, and their ability to convey trust and confidence i.e. friendly
staff, good advice, product knowledge and after sales service.
Of all the assurance attributes, respondents only perceived friendly staff to be of some
importance (slightly above average), whilst good advice, product knowledge and after
sales service was of little importance.
- 50 -
Contrary to the proposition which suggests that assurance is the second most
important service attribute in retail banking, the results show that assurance is only
ranked fourth in an African context.
Dimension Empathy
The underlying service dimension empathy reflects issues relating to the caring,
individualised attention provided to the customer; i.e. understanding the customers
personal banking needs, and building long term relationships.
Understanding personal banking needs was the only empathy attribute to be of some
importance.
Once again, the relative importance of this dimension was not consistent with previous
research. The literature review showed that empathy is typically ranked third in a retail
banking environment. The results indicate that in an African context, this dimension is
the least important, ranked fifth by users of retail banking services.
Country-specific results
When analysing the results by country (Table 11), clear similarities and differences
emerge. Efficient staff was rated as the most important service attribute for all ten
countries. Thereafter, the second and third most important service attributes varied
significantly by country.
- 51 -
Table 11: Importance ranking of standardised means, by country
Overall
Service Attributes Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim Mean
Efficient staff 1.101 0.988 0.800 0.779 0.518 0.882 0.864 0.836 0.905 0.769 0.844
Responsiveness
Shorter queues 0.416 0.925 0.403 0.249 0.409 0.259 0.544 0.278 0.245 0.399 0.413
Provides feedback on processes -0.178 -0.204 -0.241 -0.190 -0.169 -0.044 -0.004 -0.334 -0.585 0.089 -0.186
Feedback is provided timeously -0.181 -0.716 -0.167 -1.023 -0.534 -0.838 0.059 0.109 -0.446 -0.269 -0.401
Response time to info requests -0.066 0.428 0.165 0.322 0.303 0.510 -0.119 0.094 0.037 0.167 0.184
Reliability
Reliability 0.419 0.066 0.234 0.761 0.114 -0.022 0.097 0.342 0.422 0.610 0.304
Appearance of facilities -0.357 -0.477 -0.303 -0.250 -0.173 -0.370 -0.168 -0.375 -0.326 -0.571 -0.337
Tangibles
Convenient locations 0.015 0.161 0.354 0.481 -0.065 -0.068 0.285 0.475 0.419 0.318 0.237
Provides a welcoming environment -0.487 -0.810 -0.447 -0.542 -0.177 -0.266 -0.299 -0.672 -0.288 -0.799 -0.479
Friendly staff 0.022 0.117 0.586 0.069 0.299 0.019 0.480 0.203 0.191 -0.141 0.184
Assurance
After sales service -0.313 0.048 -0.360 -0.282 0.043 0.513 -0.453 -0.343 -0.201 -0.165 -0.151
Provides good advice -0.232 -0.348 -0.255 -0.166 0.063 0.269 -0.399 -0.324 -0.172 -0.318 -0.188
Good product knowledge -0.040 -0.309 -0.440 -0.677 -0.547 -0.766 -0.364 -0.404 -0.359 -0.546 -0.445
Empathy
Building long term relationships -0.356 -0.222 -0.249 0.088 -0.257 -0.254 -0.224 -0.190 0.030 0.018 -0.161
Understands personal banking needs 0.236 0.352 -0.080 0.383 0.173 0.176 -0.299 0.304 0.127 0.438 0.181
Note:
The higher the mean, the more important that aspect of service delivery is to the respondents
Negative means are inconsequential since they represent a less than average importance
Positive means are represented in black. Negative means are represented in grey.
The data has been standardised, therefore the overall mean = 0, and the overall standard
deviation = 1
The overall mean was derived from the individual country means
No weighting was applied to the overall mean in order to give each country equal importance
In summary, responsiveness was the most important dimension for Africa overall, driven
by staff efficiency and shorter queues. The results also suggest that relational issues
surrounding assurance and empathy are of less importance in an African context.
The key attributes identified (staff efficiency, shorter queues, and reliability) tend to be
more outcome based and can be classified as objective hard issues since they can be
much more readily identified and specified to that of soft and subjective relational
issues. This implies that addressing the hard objective issues could more readily provide,
and monitor, a high quality service in a fairly prescriptive and reproducible manner.
(Blanchard & Galloway, 1994)
- 52 -
Furthermore, cross-country analysis shows partial indifferences in the relative
importance regarding dimensions and attributes. Staff efficiency/ responsiveness is the
most important attribute/ dimension for each and every country, indicating that partial
similarities of relative importance across countries do exist.
This would suggest that efforts to increase speed of processing information and
customers are likely to have an important and positive effect on customer satisfaction.
Hypothesis t-tests were used to determine the significance of each attribute, for the
overall mean and for each country. The tests show if there is a significant difference from
0 for each attribute. If there is, then the attribute is significantly above average
importance and thus should be taken into consideration.
The results are based on the assumptions that each population is approximately
normally distributed and there are equal population variances. Central Limit theorem can
be used as the basis for normality due to large sample sizes. The samples were
independent and selected at random. A one tail t-test was used at a significance level of
5%. Therefore, if the p value is less than 0.05, the zero hypothesis is rejected, and the
attribute is of significant importance. These results are presented in Table 12, where
significant attributes are represented in black, and insignificant attributes are represented
in grey.
For the overall mean of each service attribute, the findings in Table 12 show that
only seven (out of fifteen) service attributes proved to be of significant importance
(p < 0.05). These were:
efficient staff
shorter queues
fast response times
service reliability
convenient locations
- 53 -
friendly staff
understanding personal banking needs
When delineating what the similarities in significant attributes are across the ten
countries, only efficient staff and shorter queues were significant for each and
every country, based on the low p values for these two attributes in Table 12.
On average, the countries tended to only have 6 significant attributes, out of the
15 total attributes.
- 54 -
Table 12: Significance of Attribute Means by Country, Q4 2006
Shorter queues 0.416 0.000 0.925 0.000 0.403 0.000 0.249 0.000 0.409 0.000
Provides feedback on processes -0.178 NA -0.204 NA -0.241 NA -0.190 NA -0.169 NA
Feedback is provided timeously -0.181 NA -0.716 NA -0.167 NA -1.023 NA -0.534 NA
Response time to info requests -0.066 NA 0.428 0.000 0.165 0.000 0.322 0.000 0.303 0.000
Reliability
Reliability 0.419 0.000 0.066 0.092 0.234 0.000 0.761 0.000 0.114 0.010
NA NA NA NA NA
Convenient locations 0.015 0.344 0.161 0.001 0.354 0.000 0.481 0.000 -0.065 NA
Provides a welcoming environment -0.487 NA -0.810 NA -0.447 NA -0.542 NA -0.177 NA
Friendly staff 0.022 0.279 0.117 0.009 0.586 0.000 0.069 0.032 0.299 0.000
Assurance
After sales service -0.313 NA 0.048 0.164 -0.360 NA -0.282 NA 0.043 0.191
Provides good advice -0.232 NA -0.348 NA -0.255 NA -0.166 NA 0.063 0.101
Good product knowledge -0.040 NA -0.309 NA -0.440 NA -0.677 NA -0.547 NA
Empathy
Building long term relationships -0.356 NA -0.222 NA -0.249 NA 0.088 0.009 -0.257 NA
Understands personal banking needs 0.236 0.000 0.352 0.000 -0.080 NA 0.383 0.000 0.173 0.000
Efficient staff 0.882 0.000 0.864 0.000 0.836 0.000 0.905 0.000 0.769 0.000
Responsiveness
Shorter queues 0.259 0.000 0.544 0.000 0.278 0.000 0.245 0.000 0.399 0.000
Provides feedback on processes -0.044 NA -0.004 NA -0.334 NA -0.585 NA 0.089 0.014
Feedback is provided timeously -0.838 NA 0.059 0.149 0.109 0.016 -0.446 NA -0.269 NA
Response time to info requests 0.510 0.000 -0.119 NA 0.094 0.031 0.037 0.191 0.167 0.000
Reliability
Reliability -0.022 NA 0.097 0.045 0.342 0.000 0.422 0.000 0.610 0.000
NA NA NA NA NA
Convenient locations -0.068 NA 0.285 0.000 0.475 0.000 0.419 0.000 0.318 0.000
Provides a welcoming environment -0.266 NA -0.299 NA -0.672 NA -0.288 NA -0.799 NA
Friendly staff 0.019 0.383 0.480 0.000 0.203 0.000 0.191 0.000 -0.141 NA
Assurance
Building long term relationships -0.254 NA -0.224 NA -0.190 NA 0.030 0.237 0.018 0.327
Understands personal banking needs 0.176 0.003 -0.299 NA 0.304 0.000 0.127 0.001 0.438 0.000
Note assumptions:
One tail t-test, Significance level: 5% ( = 0.05). X bar = mean.
Significant attributes are represented in black. Insignificant attributes are represented in grey
NA (Not Applicable): The P value results relating to negative means would imply the attribute is
insignificant and therefore not applicable, since we are testing for above average (0) importance
using a one tail t-test.
The data has been standardised, therefore the overall mean = 0, and the overall standard
deviation = 1
- 55 -
The above results highlight some of the key determinants of service quality in Africa, and
provide managers of multinational companies with a framework of similarities that exist
across countries when assessing service quality.
Based on Table 12, the significant service attributes and respective service
dimensions are shown graphically for each country, in Figure 10 to Figure 19 below,
(ranked by dimensional importance).
Kenya and Zambia followed the same service dimensional ranking as the overall
ranking for Africa i.e. (1) responsiveness, (2) reliability, (3) tangibles, (4) assurance,
and (5) empathy.
Ghana and Tanzania were the only other two countries displaying an identical
dimensional ranking i.e.: (1) responsiveness, (2) assurance, (3) tangibles, (4)
reliability. None of the African countries followed the full ranking proposed in the
second proposition: (1) responsiveness, (2) assurance, (3) empathy, (4) tangibles, (5)
reliability.
While partial similarities are evident, it is clear that cross-national differences across
most countries are prevalent. Furthermore, responsiveness being the most important
dimension is the only similarity between African countries and the UK.
- 56 -
Figure 10: Significant Service Dimensions and Attributes for Botswana
1.200
1.000
0.800
Mean
0.600
0.400
0.200
0.000
Understands
queues
Reliability
Efficient
Shorter
staff
personal
banking
needs
Responsiveness Reliability Empathy
1.200
1.000
0.800
Mean
0.600
0.400
0.200
0.000
Understands
time to info
queues
Friendly
Efficient
Convenient
Shorter
Response
requests
locations
staff
personal
staff
banking
needs
- 57 -
Figure 12: Significant Service Dimensions and Attributes for Ghana
0.900
0.800
0.700
0.600
0.500
Mean
0.400
0.300
0.200
0.100
0.000
time to info
queues
Friendly
Reliability
Efficient
Convenient
Shorter
Response
requests
locations
staff
staff
Responsiveness Assurance Tangibles Reliability
0.900
0.800
0.700
0.600
0.500
Mean
0.400
0.300
0.200
0.100
0.000
Understands
time to info
queues
Reliability
relationships
Efficient staff
Convenient
Friendly staff
Building long
Shorter
Response
requests
locations
personal
banking
needs
term
- 58 -
Figure 14: Significant Service Dimensions and Attributes for Mauritius
0.600
0.500
0.400
Mean
0.300
0.200
0.100
0.000
Understands
time to info
queues
Friendly
Efficient
Reliability
Shorter
Response
requests
staff
personal
staff
banking
needs
Responsiveness Assurance Reliability Empathy
1.000
0.900
0.800
0.700
0.600
Mean
0.500
0.400
0.300
0.200
0.100
0.000
Understands
time to info
queues
After sales
Efficient
good advice
Shorter
Response
requests
Provides
staff
service
personal
banking
needs
- 59 -
Figure 16: Significant Service Dimensions and Attributes for Tanzania
1.000
0.900
0.800
0.700
0.600
Mean
0.500
0.400
0.300
0.200
0.100
0.000
queues
Friendly
Reliability
Efficient
Convenient
Shorter
locations
staff
staff
Responsiveness Assurance Tangibles Reliability
0.900
0.800
0.700
0.600
0.500
Mean
0.400
0.300
0.200
0.100
0.000
Understands
time to info
queues
Reliability
Friendly
Feedback is
Efficient
Convenient
Shorter
Response
requests
timeously
locations
staff
provided
personal
staff
banking
needs
- 60 -
Figure 18: Significant Service Dimensions and Attributes for Zambia
1.000
0.900
0.800
0.700
0.600
Mean
0.500
0.400
0.300
0.200
0.100
0.000
Understands
queues
Efficient
Reliability
Friendly
Convenient
Shorter
locations
staff
personal
staff
banking
needs
Responsiveness Reliability Tangibles Assurance Empathy
0.900
0.800
0.700
0.600
0.500
Mean
0.400
0.300
0.200
0.100
0.000
Understands
time to info
queues
Reliability
Efficient
Convenient
feedback on
Shorter
Response
processes
requests
locations
Provides
staff
personal
banking
needs
- 61 -
4.4 Changes in Service Expectations over Time (Q1 and Q4 2006)
In order to determine whether customer expectations for retail banking services in Africa
are constant over time, hypothesis t-testing was used to test for the significance of
differences between country means over time, comparing data from Q1 2006 with Q4
2006.
The results are based on the assumptions that each population for both data sets are
approximately normally distributed and there are equal population variances. Central
Limit theorem can be used as the basis for normality due to large sample sizes. The
samples were independent and selected at random for both dips. A two tail t-test was
conducted at a 5% level of significance. If the p value is less than 0.05, the zero
hypothesis is rejected, and there is a significant change in sample means over time.
These results are shown in Table 13 and 14.
Analysis of the overall means (Table 13) show that the Ho is rejected for ten out of
eleven attributes based on a 5% level of significance. This implies that significant
changes in (mean) importance from Q1 2006 to Q4 2006 were evident across almost all
attributes. The only attribute that remained consistent over time was understands
personal banking needs.
Therefore, there is sufficient reason to believe that customer service expectations in the
African retail banking sector do change over time, and that contrary to the statement by
deCarvalho & Leite (1999), these changes are evident over a relatively short period of
time (i.e. less than one year).
- 62 -
Table 13: T-test for Significance of differences between overall service attribute
means over time (Q1 - Q4 2006)
Overall
Service Attributes Means
x P Value
1. Efficient staff Q4 2006 0.844 0.000
Q1 2006 0.691
2. Shorter queues Q4 2006 0.413 0.000
Q1 2006 0.322
3. Convenient locations Q4 2006 0.237 0.000
Q1 2006 -0.041
4. Friendly staff Q4 2006 0.184 0.000
Q1 2006 0.067
5. After sales service Q4 2006 -0.151 0.000
Q1 2006 -0.307
6. Response time to info requests Q4 2006 0.184 0.008
Q1 2006 0.119
7. Reliability Q4 2006 0.304 0.000
Q1 2006 0.186
8. Appearance of facilities Q4 2006 -0.337 0.000
Q1 2006 -0.520
9. Building long term relationships Q4 2006 -0.161 0.000
Q1 2006 -0.384
10. Understands personal banking needs Q4 2006 0.181 0.053
Q1 2006 0.134
11. Provides good advice Q4 2006 -0.188 0.002
Q1 2006 -0.265
Note:
Two tail-t test was used
Significance level: 5% ( = 0.05)
The data has been standardised, therefore the overall mean = 0, and the overall standard
deviation = 1
The overall mean was derived from the individual country means
No weighting was applied to the overall mean in order to give each country equal importance
For efficient staff, only four out of ten countries displayed a significant difference
between country means over time, which tells us that for most countries, efficient staff
was consistently perceived to be the most important attribute over time. It is also
- 63 -
interesting to note however that of those four countries that showed a significant change,
an upward mean trend for all four countries was evident i.e. having efficient staff is
becoming increasingly important for retail banking customers in Africa.
A similar trend was true for the attribute shorter queues, where only four out of ten
countries showed a significant difference in mean importance over time; however most of
these changes displayed a tendency towards increasing importance for shorter queues.
Egypt and Kenya were the most volatile countries regarding changes in service
expectations, representing the highest number of attributes (i.e. 8) that changed
significantly over time. This is most likely linked to changes in customer experience and
attitude (Woodruff, 1997). Conversely, Seychelles represented the least change (3
attributes) in service expectations over time.
Table 14: T-test for significance of differences between country attribute means
over time (Q1 - Q4 2006)
- 64 -
Service Attributes Seych Tanz Ugan Zam Zim
x x P Value x P Value x P Value x P Value x P Value
1. Efficient staff Q4 2006 0.882 0.098 0.864 0.000 0.836 0.000 0.905 0.453 0.769 0.140
Q1 2006 0.678 0.383 0.472 0.959 0.872
2. Shorter queues Q4 2006 0.259 0.217 0.544 0.555 0.278 0.003 0.245 0.035 0.399 0.745
Q1 2006 0.107 0.599 0.005 0.397 0.377
3. Convenient locations Q4 2006 -0.068 0.001 0.285 0.000 0.475 0.000 0.419 0.001 0.318 0.003
Q1 2006 -0.486 -0.222 0.063 0.184 0.110
4. Friendly staff Q4 2006 0.019 0.067 0.480 0.000 0.203 0.741 0.191 0.136 -0.141 0.018
Q1 2006 0.246 -0.170 0.174 0.083 0.024
5. After sales service Q4 2006 0.513 0.712 -0.453 0.004 -0.343 0.153 -0.201 0.000 -0.165 0.000
Q1 2006 0.467 -0.177 -0.472 -0.499 -0.568
6. Response time to info requests Q4 2006 0.510 0.081 -0.119 0.000 0.094 0.702 0.037 0.050 0.167 0.000
Q1 2006 0.295 0.251 0.129 -0.104 -0.083
7. Reliability Q4 2006 -0.022 0.157 0.097 0.220 0.342 0.113 0.422 0.047 0.610 0.814
Q1 2006 -0.197 -0.019 0.486 0.279 0.594
8. Appearance of facilities Q4 2006 -0.370 0.027 -0.168 0.088 -0.375 0.130 -0.326 0.001 -0.571 0.052
Q1 2006 -0.644 -0.329 -0.238 -0.561 -0.706
9. Building long term relationships Q4 2006 -0.254 0.000 -0.224 0.005 -0.190 0.207 0.030 0.000 0.018 0.000
Q1 2006 -0.774 0.044 -0.304 -0.403 -0.350
10. Understands personal banking needs Q4 2006 0.176 0.825 -0.299 0.000 0.304 0.000 0.127 0.548 0.438 0.000
Q1 2006 0.203 0.146 -0.075 0.084 -0.010
11. Provides good advice Q4 2006 0.269 0.183 -0.399 0.258 -0.324 0.355 -0.172 0.001 -0.318 0.390
Q1 2006 0.105 -0.506 -0.240 -0.419 -0.258
Note assumptions:
Two tail t-test
Significance level: 5% ( = 0.05)
The data has been standardised, therefore the overall mean = 0, and the overall standard
deviation = 1
Thus we can deduce that while no changes were evident in the relative importance
ranking of the top three attributes over time, i.e. (efficient staff, shorter queues, and
reliability), results from the t-test show significant changes in (mean) importance between
Q1 2006 to Q4 2006 for ten out of eleven attributes. Furthermore, these changes are
evident in service expectations over a short term period. (i.e. 8 months). Egypt and
Kenya demonstrated the highest level of volatility in expectations. Of interest, efficient
staff, shorter queues and convenient locations are becoming increasingly important for
customers. This calls for continuous short term assessments of changing customer
expectations. This will help managers concentrate their efforts in service quality
improvements where most beneficial.
The following chapter presents an interpretation of the results and an analysis of the
possible implications for management wishing to improve the quality of service delivery.
- 65 -
5 CONCLUSIONS AND RECOMMENDATIONS
The objective of this research was to analyse customer service expectations in retail
banking in Africa. In order to assess this, propositions based on the research sub-
problems and on the literature review were formulated as follows:
The following conclusions were derived from this research for the respective
propositions:
The research clearly shows that customer service expectations in retail banking do differ
significantly between countries in Africa: Accept Proposition One.
ANOVA testing presents sufficient evidence to indicate that for every service
attribute, one or more of the population (country) means are not equal to the
others. Every attribute had some significant difference between countries. This
implies that different cultural groups give different importance to service quality
dimensions.
- 66 -
It is therefore critical to take cross-national differences into consideration when
designing and implementing a marketing strategy for multinational companies. In
order to achieve this, country-specific customer expectations must be determine
and prioritised as a first step towards improving service quality.
The importance ranking of the service dimensions in African retail banking are:
1. Responsiveness
2. Assurance
3. Empathy
4. Tangibles
5. Reliability
The relative importance of the underlying SERVQUAL dimensions in this survey show
partial variance to those identified by past research: Partially Accept Proposition Two.
1. Responsiveness 1. Responsiveness
2. Assurance 2. Reliability
3. Empathy 3. Tangibles
4. Tangibles 4. Assurance
5. Reliability 5. Empathy
- 67 -
This assessment is based on the ranking of individual attributes collapsed into the
underlying dimensions. The data from the survey is derived from a partial rank
ordering technique. Rescaling of the results using the Stochastic Search
Algorithm (Stacey, 2006) converted the results into standardised sample means,
in order to compare mean rankings.
This research also identifies the individual attributes that collectively drive the
underlying dimensions. Staff efficiency is not only the most important attribute (on
average) for retail banking customers in Africa, but it is perceived to be twice as
important as having shorter queues, ranked second. Service reliability and
convenient locations are rated as the third and fourth most important attributes
respectively (out of 15 attributes in total). Of least importance is having a
welcoming environment and good product knowledge.
- 68 -
Table 16: Importance Ranking of Service Attributes
Ranking: by country
Results from the t-tests used to determine the significance of attributes indicate
that only seven (out of fifteen) service attributes proved to be of significant
importance (i.e. significantly above average importance). These were efficient
staff, shorter queues, fast response times, service reliability, convenient locations,
friendly staff, and understanding personal banking needs.
- 69 -
Significant attributes: by country
When delineating what the similarities in significant attributes are across the ten
countries, only efficient staff and shorter queues were significant for each and
every country. Appearance of facilities, providing a welcoming environment and
good product knowledge proved to be insignificant attributes for all ten countries.
Any time and money put into these areas (over and above the standard offering)
might be better redirected elsewhere.
The above results highlight some of the key determinants of service quality in
Africa, and provide managers of multinational companies with a framework of
similarities that exist across countries. If detailed branch analysis is not available,
this research highlights two general areas that banks need to focus their attention
on in order to achieve an advantage.
Kenya and Zambia followed the same dimensional ranking as the overall ranking
for Africa i.e. (1) responsiveness, (2) reliability, (3) tangibles, (4) assurance, (5)
empathy. Ghana and Tanzania were the only other two countries displaying an
identical dimensional ranking i.e.: (1) responsiveness, (2) assurance, (3)
tangibles, and (4) reliability. None of the African countries followed the full ranking
proposed in the second proposition, i.e. (1) responsiveness, (2) assurance, (3)
empathy, (4) tangibles, and (5) reliability.
- 70 -
5.1.3 Research Proposition 3:
Customer service expectations in the African retail banking sector change over
time
This survey shows that while there were no changes in the ranking of customer service
expectations over time, there were significant changes in the importance of these
expectations. Partially Accept Proposition Three.
While no changes were observed in the relative importance ranking of the top
three attributes over time, i.e. (efficient staff, shorter queues, and reliability),
results from the t-test show significant changes in (mean) importance between Q1
2006 to Q4 2006 were evident for ten out of eleven attributes. Thus we can
conclude that not only are there significant changes in customer service
expectations over time, but that significant changes are evident in service
expectations over a short term period. (i.e. 8 months). Egypt and Kenya
demonstrated the highest level of volatility in expectations, while Seychelles
remained relatively consistent for the period.
- 71 -
Table 18: Significant Changes in Mean Importance between Q1 2006 Q4 2006
Overall
Changes in Mean Importance: Means
(Q1 and Q4 2006)
x P Value
Efficient staff Q4 2006 0.844 0.000
Q1 2006 0.691
Shorter queues Q4 2006 0.413 0.000
Q1 2006 0.322
Convenient locations Q4 2006 0.237 0.000
Q1 2006 -0.041
Friendly staff Q4 2006 0.184 0.000
Q1 2006 0.067
After sales service Q4 2006 -0.151 0.000
Q1 2006 -0.307
Response time to info requests Q4 2006 0.184 0.008
Q1 2006 0.119
Reliability Q4 2006 0.304 0.000
Q1 2006 0.186
Appearance of facilities Q4 2006 -0.337 0.000
Q1 2006 -0.520
Building long term relationships Q4 2006 -0.161 0.000
Q1 2006 -0.384
Understands personal banking needs Q4 2006 0.181 0.053
No significant
Q1 2006 0.134 change
Provides good advice Q4 2006 -0.188 0.002
Q1 2006 -0.265
Of interest, efficient staff, shorter queues and convenient locations are becoming
increasingly important for customers.
- 72 -
5.2 Recommendations to Organisations
The results of this research provide a number of clear implications for the management
of customer service expectations.
However some similarities are evident for certain dimensions and attributes, which
can be leveraged off for a pan African strategy.
Customers of retail banks in Africa are clearly seeking a responsive service, with a
high level of staff efficiency. This would suggest that efforts to increase speed of
processing information and customers are likely to have an important and positive
effect on customer satisfaction. Addressing this issue through regular staff training
should improve service quality.
Shorter queues, service reliability and convenient locations are also considered to
be very important by customers. Furthermore, these attributes (including staff
efficiency) are becoming increasingly important over time. For retail banks in
Africa wanting to enhance customer satisfaction, these would be key areas for
improvement. These are areas where well-designed routines and responses could
be used to gain maximum impact.
Service expectations in Africa do change over time, and are evident over a
relatively short period of time (< 1 year). It is recommended that short term
assessments of changes in customer expectations are done in order to accurately
meet customer needs and improve service quality.
- 73 -
5.3 Recommendations for Future Research
Researchers should also consider taking a broader view towards identifying the
components of service quality and the overall service offering. The components should
encompass not only the recently identified external constructs relating to customer
expectations, but also the internal issues relating to employee satisfaction that drive
service quality (i.e. the full service profit chain).
Future research of a case study nature into the key service attributes delivered by retail
banks in Africa would provide valuable data for the improvement of the questionnaire
format.
- 74 -
REFERENCES
Anderson, E.A. & Sullivan, M.W. (1993), "The antecedents and consequences of
customer satisfaction for firms", Marketing Science, Vol. 12, pp. 125-43.
Berry, L.L., Zeithaml, V.A., Parasuraman, A. (1985), "Quality counts in services, too",
Business Horizons, May-June, pp.44-52.
Bick, G., Brown A.G. & Abratt R. (2004), Customer perceptions of the value delivered by
retail banks in South Africa, The International Journal of Bank Marketing, Vol. 22 (4/5);
pg. 300
Bitner, M.J., Booms, B.H., Tetreault, M.S. (1990), "The service encounter: diagnosing
favorable and unfavorable incidents", Journal of Marketing, Vol. 54 pp.71-84.
Bitner, M.J., & Hubbert, A.R. (1994), "Encounter satisfaction versus overall satisfaction
versus quality", in Rust, R.T. & Oliver, R.L. (Eds), Service Quality: New Directions in
Theory and Practice, Sage, London.
Blanchard, R.F & Galloway, R.L (1994), Quality in Retail Banking, International Journal
of Service Industry Management, Vol. 5 (4), pp. 5-23.
- 75 -
Boulding, W., Kalra, A., Staelin, R. & Zeithaml, V.A. (1993), A Dynamic Process Model
of Service Quality: From Expectations to Behavioural Intentions, Journal of Marketing
Research, Vol. 30 (1), pp. 7-27.
Cina, C. (1990), "Five steps to service excellence", The Journal of Services Marketing,
Vol. 4 pp. 39-47.
Cronin, J.J. Jr. & Taylor, S.A. (1992), Measuring Service Quality: a re-examination and
extension, Journal of Marketing, Vol. 56 (3), pp. 55-69
Dabholkar, P.A., Thorpe, D.I. & Rentz, J.O. (1996), "A measure of service quality for
retail stores: scale development and validation", Journal of the Academy of Marketing
Science, Vol. 24, Winter, pp. 3-16.
Daniel, A.L. (1992), "Overcome the barriers to superior customer service", Business and
Society Review, pp. 53-5.
Davidow, W.H. & Uttal, B. (1989), Service companies: focus or falter, Harvard Business
Review, July/August, pp. 17-34.
DeCarvalho, F.A. & Leite, V.F. (1999), Attribute Importance in Service Quality: An
empirical test of the PZB conjecture in Brazil, International Journal of Service Industry
Management, Vol.10 (5), pp. 487-499
- 76 -
Denzin, N.K., & Lincoln, Y.S. (1998), The landscape of qualitative research: theories and
issues, Thousand Oaks, California, Sage Publications.
File, K.M. & Prince, R.A. (1992), "Positive word of mouth: customer satisfaction and
buyer behaviour", International Journal of Bank Marketing, Vol. 10 (1), pp. 25-9.
Gale, B.T. (1994), Managing Customer Value, Free Press, New York, NY.
Grnroos, C. (1988), Service quality: The six criteria of good perceived service, Review
of Business, Vol. 9 (3), pp.10-14
Heskett, J.L., Jones, T.O., Loveman, G.W., Sasser, W.E. Jr, & Schlesinger, L.A. (1994),
Putting the Service-Profit Chain to Work, Harvard Business Review, March-April, pp.
164-74
Heskett, J.L., Sasser, W.E., & Schlesinger, L.A. (1997), The Service Profit Chain, New
York: Free Press, pp 302
- 77 -
Johnston, R. (1997), Identifying the critical determinants of service quality in retail
banking: importance and effect, International Journal of Bank Marketing, Vol. 15 (4), pp.
111-116.
Kamakura, W.A., Mittal, V., De Rosa, F., & Mazzon, J. (2002), Assessing the Service-
Profit Chain, Marketing Science; Vol. 21 (3), pp. 294-317
Keiningham, T., Zahorik, A.J., & Rust, R.T. (1995), Getting return on equity, Journal of
Retail Banking, 16 (4), 7-13.
Leedy, P.D. & Ormrod, J.E. (2001), Practical Research: Planning and Design, Seventh
Edition, Upper Saddle River: Merrill Prentice Hall.
- 78 -
Marshall, G.W., Baker, J., & Finn D.W. (1998), Exploring internal customer service
quality, Journal of Business & Industrial Marketing, Vol. 13 (4/5) pp. 381.
Nel, D., Pitt, L.F. & Berthon, P.R. (1997), The SERVQUAL Instrument: reliability and
validity in South Africa, South African Journal of Business Management, Vol. 28 (3), pp.
113-122.
Newman, K. & Cowling, A. (1996), "Service quality in retail banking: the experience of
two British clearing banks", International Journal of Bank Marketing, Vol. 14, pp. 3-11.
Parasuraman, A., Zeithaml, V.A., & Berry, L.L. (1985), A Conceptual Model of Service
Quality and its Implications for Future Research, Journal of Marketing, Vol. 58, January,
pp. 111-24.
Parasuraman, A., Zeithaml, V.A., & Berry, LL. (1985), "Guidelines for Conducting Service
Quality Research, Marketing Research, December, pp. 33-34.
Parasuraman, A., Berry, LL., & Zeithaml, V.A. (1985), "A conceptual model of service
quality and its implications for future research", Journal of Marketing, Vol. 49, pp. 41-50.
Parasuraman, A., Zeithaml, V.A., & Berry, L.L. (1988), SERVQUAL: A Multiple-Item
Scale for Measuring Consumer Perceptions of Service Quality, Journal of Retailing, Vol.
64 (1), 12-40
Parasuraman, A., Zeithaml, V.A., & Berry, LL. (1990), "A Conceptual Model of Service
Quality and its Implications for Future Study, Journal of Marketing, Vol. 49 (Fall), pp. 41-
50.
- 79 -
Parasuraman, A., Berry, L.L., & Zeithaml, V.A. (1990), Delivering Quality Service, The
Free Press, New York, NY, Ch 2.
Parasuraman, A., Berry, L.L., & Zeithaml, V.A. (1991), Understanding Customer
Expectations of Service, Sloan Management Review; ABI/INFORM Global; pp 39.
Parasuraman, A., Zeithaml, V.A., & Berry, LL. (1993), "Refinement and Reassessment of
the SERVQUAL Scale, Journal of Retailing, Vol. 69 (Spring), pp. 140-147.
Parasuraman, A., Zeithaml, V.A., & Berry, L.L. (1994), Reassessment of Expectations
as a Comparison Standard in Measuring Service Quality: Implication for further
research, Journal of Marketing, Vol. 58 (1), pp 111-125.
Reichheld, F.F. & Sasser, W.E. Jr (1990), "Defections: quality comes to services", The
Harvard Business Review, pp. 106-7.
Richens, M.L. (1983), "Negative word of mouth by dissatisfied consumers: a pilot study",
Journal of Marketing, Winter, p. 69.
Robledo, M.A. (2001), Measuring and Managing Service Quality: Integrating Customer
Expectations, Managing Service Quality, Vol. 11 (1), pp. 22-31.
Rucci, A.J. Kirn, S.P., & Quinn, R.T. The employee-customer-profit chain at Sears.
Harvard Business Review (Jan-Feb) pp 83-97.
- 80 -
Rust, R.T. & Oliver, R.L. (1994), Service Quality: New Dimensions in Theory and
Practice, Sage, London.
Rust, R.T., Zahorik A.J., & Keiningham, T.L. (1995), Return on Quality (ROQ): Making
Service Quality Financially Accountable, Journal of Marketing, Vol. 59 (Apr), pp 58-70
Shycon, H.N. (1992), "Improving Customer Service: Measuring the Payoff", The Journal
of Business Strategy, pp. 13-17.
Smith, A.M. & Reynolds, N.L. (2001), Measuring Cross-National Service Quality - A
Framework for assessment, International Marketing Review, Vol. 19 (5), pp. 450-481.
Stacey, A. (2006), Estimating the Means and Standard Deviations of Rank Ordered
Survey Items, Journal of Management Dynamics, Vol. 15 (3), pp1-14.
Stone, J.R. (2006), Assessing the Quality of Service Delivered by South African
Consulting Engineers, Unpublished MBA Research Report, Johannesburg: University of
the Witwatersrand.
Taylor, S.A. & Baker, T.L. (1994), "An Assessment of the Relationship between Service
Quality and Customer Satisfaction in the Formation of Consumers' Purchase Intentions",
Journal of Retailing, Vol. 70 (2), pp. 163-78.
- 81 -
Tenner, A.R. & DeToro, I.J. (1992), Total Quality Management, Addison-Wesley
Publishing Co, Inc., Reading, MA.
Venetis, K.A and Ghauri, P.N. (2004), Service quality and Customer retention: building
long-term relationships, European Journal of Marketing, Vol. 38 (11/12), pp.1577-1593.
Webster, C. (1989), Can consumers be segmented on the basis of their service quality
expectations?, Journal of Services Marketing, Vol. 2 (3), pp. 35-53.
Woodruff, R.B. (1997), Customer value: The next source of competitive advantage,
Journal of the Academy of Marketing Science, Vol. 25 (2), pp.139-154
World Trade Organisation, (2005), International trade statistics 2005, By Sector (Pdf)
http://www.wto.org/english/res_e/statis_e/its2005_e/its05_bysector_e.pdf
Zeithaml, V.A., Parasuraman, A., & Berry, L.L. (1990), Delivering Quality Service, The
Free - Press, New York, NY.
- 82 -
Zeithaml, V., Berry, L.L., & Parasuraman, A. (1993), The nature and determinants of
customer expectations of a service, Journal of the Academy of Marketing Science, Vol.
21 (1), pp. 1-12.
- 83 -
APPENDIX A: INTERVIEW QUESTIONNAIRES
1. I am going to read to you a list of service areas relating to your bank. After I read the
whole list (there are eleven service areas) please say which three service areas are most
important to you and place them in order of First Most Important, Second Most Important
and Third Most Important.
(Interviewer note: The respondent can only answer after you read the whole list. Tick one block only for
First Most Important, tick one block only for Second Most Important and tick one block only for Third
Most Important.)
1. Efficient staff
2. Shorter queues
3. Convenient locations
4. Friendly Staff
8. Appearance of facilities
- 84 -
14. Feedback is provided timeously on key
steps in processes.
E.g. loan approvals or account openings
1. I am going to read to you a list of service areas relating to your bank. After I read the
whole list (there are eleven service areas) please say which three service areas are most
important to you and place them in order of First Most Important, Second Most Important
and Third Most Important.
(Interviewer note: The respondent can only answer after you read the whole list. Tick one block only for
First Most Important, tick one block only for Second Most Important and tick one block only for Third
Most Important.)
1. Efficient staff
2. Shorter queues
3. Convenient locations
4. Friendly Staff
8. Appearance of facilities
- 85 -
APPENDIX B: DETAILED PRESENTATION OF RESULTS
- 86 -
Summary stats for Q4 2006 ITEM 5: GOOD AFTER SALES SERVICE
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.313 0.048 -0.360 -0.282 0.043 0.513 -0.453 -0.343 -0.201 -0.165
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
Summary stats for Q4 2006 ITEM 6: FAST RESPONSE TIME TO INFO REQUESTS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.066 0.428 0.165 0.322 0.303 0.510 -0.119 0.094 0.037 0.167
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
Summary stats for Q4 2006 ITEM 9: FOCUS ON BUILDING LONG TERM RELATIONSHIPS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.356 -0.222 -0.249 0.088 -0.257 -0.254 -0.224 -0.190 0.030 0.018
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
- 87 -
Summary stats for Q4 2006 ITEM 10: UNDERSTANDS MY PERSONAL BANKING NEEDS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means 0.236 0.352 -0.080 0.383 0.173 0.176 -0.299 0.304 0.127 0.438
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
Summary stats for Q4 2006 ITEM 13: PROVIDES FEEDBACK ON KEY STEPS IN PROCESSES eg. Loan Appro
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.178 -0.204 -0.241 -0.190 -0.169 -0.044 -0.004 -0.334 -0.585 0.089
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
Summary stats for Q4 2006 ITEM 14: PROMISED FEEDBACK IS PROVIDED TIMEOUSLY
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.181 -0.716 -0.167 -1.023 -0.534 -0.838 0.059 0.109 -0.446 -0.269
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
- 88 -
Summary stats for Q4 2006 ITEM 15: STAFF HAVE GOOD KNOWLEDGE OF THE FULL PRODUCT RANGE
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 560 316 496 571 380 200 248 330 488 494
Sample means -0.040 -0.309 -0.440 -0.677 -0.547 -0.766 -0.364 -0.404 -0.359 -0.546
Sample standard deviations 0.908 0.875 0.925 0.884 0.953 0.911 0.894 0.918 0.938 0.898
Sample variances 0.824 0.766 0.856 0.781 0.908 0.829 0.799 0.843 0.880 0.806
Weights for pooled variance 0.137 0.077 0.122 0.140 0.093 0.049 0.061 0.081 0.120 0.121
- 89 -
Appendix B.2: Summary Stats Q1 2006
- 90 -
Summary stats for Q1 2006 ITEM 5: GOOD AFTER SALES SERVICE
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 300 152 250 302 200 75 152 155 250 251
Sample means -0.603 -0.285 -0.348 -0.507 -0.078 0.467 -0.177 -0.472 -0.499 -0.568
Sample standard deviations 0.882 0.856 0.934 0.896 0.920 0.901 0.953 0.959 0.901 0.892
Sample variances 0.777 0.732 0.872 0.803 0.846 0.813 0.909 0.920 0.811 0.795
Weights for pooled variance 0.144 0.073 0.120 0.145 0.096 0.036 0.073 0.074 0.120 0.120
Summary stats for Q1 2006 ITEM 6: FAST RESPONSE TIME TO INFO REQUESTS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 300 152 250 302 200 75 152 155 250 251
Sample means 0.134 0.237 0.105 -0.182 0.413 0.295 0.251 0.129 -0.104 -0.083
Sample standard deviations 0.882 0.856 0.934 0.896 0.920 0.901 0.953 0.959 0.901 0.892
Sample variances 0.777 0.732 0.872 0.803 0.846 0.813 0.909 0.920 0.811 0.795
Weights for pooled variance 0.144 0.073 0.120 0.145 0.096 0.036 0.073 0.074 0.120 0.120
Summary stats for Q1 2006 ITEM 9: FOCUS ON BUILDING LONG TERM RELATIONSHIPS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 300 152 250 302 200 75 152 155 250 251
Sample means -0.292 -0.630 -0.462 -0.056 -0.611 -0.774 0.044 -0.304 -0.403 -0.350
Sample standard deviations 0.882 0.856 0.934 0.896 0.920 0.901 0.953 0.959 0.901 0.892
Sample variances 0.777 0.732 0.872 0.803 0.846 0.813 0.909 0.920 0.811 0.795
Weights for pooled variance 0.144 0.073 0.120 0.145 0.096 0.036 0.073 0.074 0.120 0.120
- 91 -
Summary stats for Q1 2006 ITEM 10: UNDERSTANDS MY PERSONAL BANKING NEEDS
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sample sizes 300 152 250 302 200 75 152 155 250 251
Sample means 0.223 0.123 -0.088 0.515 0.214 0.203 0.146 -0.075 0.084 -0.010
Sample standard deviations 0.882 0.856 0.934 0.896 0.920 0.901 0.953 0.959 0.901 0.892
Sample variances 0.777 0.732 0.872 0.803 0.846 0.813 0.909 0.920 0.811 0.795
Weights for pooled variance 0.144 0.073 0.120 0.145 0.096 0.036 0.073 0.074 0.120 0.120
- 92 -
Appendix B.3: One-Way ANOVA Testing Q4 2006
Source SS df MS F p-value
Between variation 92.520 9 10.280 12.386 0.0000
Within variation 3380.559 4073 0.830
Total variation 3473.079 4082
Source SS df MS F p-value
Between variation 125.751 9 13.972 16.834 0.0000
Within variation 3380.559 4073 0.830
Total variation 3506.310 4082
Source SS df MS F p-value
Between variation 159.799 9 17.755 21.392 0.0000
Within variation 3380.559 4073 0.830
Total variation 3540.357 4082
- 93 -
Results of one-way ANOVA
Number of samples 10
Total sample size 4083
Grand mean 0.173
Pooled variance 0.830
Pooled standard deviation 0.911
Source SS df MS F p-value
Between variation 187.733 9 20.859 25.132 0.0000
Within variation 3380.559 4073 0.830
Total variation 3568.292 4082
Source SS df MS F p-value
Between variation 190.192 9 21.132 25.461 0.0000
Within variation 3380.559 4073 0.830
Total variation 3570.751 4082
Source SS df MS F p-value
Between variation 126.368 9 14.041 16.917 0.0000
Within variation 3380.559 4073 0.830
Total variation 3506.927 4082
- 94 -
Results of one-way ANOVA
Number of samples 10
Total sample size 4083
Grand mean 0.365
Pooled variance 0.830
Pooled standard deviation 0.911
Source SS df MS F p-value
Between variation 231.092 9 25.677 30.936 0.0000
Within variation 3380.559 4073 0.830
Total variation 3611.651 4082
Source SS df MS F p-value
Between variation 56.507 9 6.279 7.565 0.0000
Within variation 3380.559 4073 0.830
Total variation 3437.066 4082
Source SS df MS F p-value
Between variation 100.798 9 11.200 13.494 0.0000
Within variation 3380.559 4073 0.830
Total variation 3481.357 4082
- 95 -
Results of one-way ANOVA
Number of samples 10
Total sample size 4083
Grand mean 0.203
Pooled variance 0.830
Pooled standard deviation 0.911
Source SS df MS F p-value
Between variation 162.351 9 18.039 21.734 0.0000
Within variation 3380.559 4073 0.830
Total variation 3542.910 4082
Source SS df MS F p-value
Between variation 102.007 9 11.334 13.656 0.0000
Within variation 3380.559 4073 0.830
Total variation 3482.566 4082
Source SS df MS F p-value
Between variation 169.220 9 18.802 22.654 0.0000
Within variation 3380.559 4073 0.830
Total variation 3549.779 4082
- 96 -
Results of one-way ANOVA
Number of samples 10
Total sample size 4083
Grand mean -0.200
Pooled variance 0.830
Pooled standard deviation 0.911
Source SS df MS F p-value
Between variation 135.587 9 15.065 18.151 0.0000
Within variation 3380.559 4073 0.830
Total variation 3516.146 4082
Source SS df MS F p-value
Between variation 499.671 9 55.519 66.891 0.0000
Within variation 3380.559 4073 0.830
Total variation 3880.230 4082
Source SS df MS F p-value
Between variation 163.099 9 18.122 21.834 0.0000
Within variation 3380.559 4073 0.830
Total variation 3543.658 4082
- 97 -
Appendix B.4: T Test: Significance of Country Means Q4 2006
Table B 4.5: T Test: Significance of Country Means, Q4 2006 After Sales Service
- 98 -
t Test: Significance of country means
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Std Sample mean - pop mean -0.313 0.048 -0.360 -0.282 0.043 0.513 -0.453 -0.343 -0.201 -0.165
Square Root of n 23.7 17.8 22.3 23.9 19.5 14.1 15.7 18.2 22.1 22.2
Std dev / SQRT n 0.038 0.049 0.042 0.037 0.049 0.064 0.057 0.051 0.042 0.040
t= -8.159 0.978 -8.672 -7.638 0.877 7.960 -7.986 -6.780 -4.743 -4.091
df = (n-1) 559 315 495 570 379 199 247 329 487 493
p-value #NUM! 0.1643 #NUM! #NUM! 0.1906 0.0000 #NUM! #NUM! #NUM! #NUM!
- 99 -
t Test: Significance of country means
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Std Sample mean - pop mean -0.356 -0.222 -0.249 0.088 -0.257 -0.254 -0.224 -0.190 0.030 0.018
Square Root of n 23.7 17.8 22.3 23.9 19.5 14.1 15.7 18.2 22.1 22.2
Std dev / SQRT n 0.038 0.049 0.042 0.037 0.049 0.064 0.057 0.051 0.042 0.040
t= -9.286 -4.504 -6.002 2.387 -5.251 -3.944 -3.939 -3.756 0.718 0.449
df = (n-1) 559 315 495 570 379 199 247 329 487 493
p-value #NUM! #NUM! #NUM! 0.0087 #NUM! #NUM! #NUM! #NUM! 0.2366 0.3268
- 100 -
t Test: Significance of country means
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Std Sample mean - pop mean -0.178 -0.204 -0.241 -0.190 -0.169 -0.044 -0.004 -0.334 -0.585 0.089
Square Root of n 23.7 17.8 22.3 23.9 19.5 14.1 15.7 18.2 22.1 22.2
Std dev / SQRT n 0.038 0.049 0.042 0.037 0.049 0.064 0.057 0.051 0.042 0.040
t= -4.632 -4.139 -5.798 -5.131 -3.449 -0.690 -0.065 -6.613 -13.778 2.215
df = (n-1) 559 315 495 570 379 199 247 329 487 493
p-value #NUM! #NUM! #NUM! #NUM! #NUM! #NUM! #NUM! #NUM! #NUM! 0.0136
- 101 -
Appendix B.5: T Test: Significance of Differences between Country
Means over Time (Q1 Q4 2006)
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 3.516 1.506 4.564 1.545 -1.658 1.660 5.095 4.005 -0.751 -1.478
p-value 0.0005 0.1328 0.0000 0.1228 0.0979 0.0980 0.0000 0.0001 0.4530 0.1399
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= -0.666 1.776 2.482 4.011 1.559 1.237 -0.591 3.011 -2.112 0.325
p-value 0.5058 0.0763 0.0133 0.0001 0.1195 0.2171 0.5548 0.0027 0.0350 0.7450
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 1.985 4.944 -1.182 4.434 3.170 3.401 5.364 4.542 3.267 3.001
p-value 0.0475 0.0000 0.2375 0.0000 0.0016 0.0008 0.0000 0.0000 0.0011 0.0028
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= -1.176 1.597 4.266 1.283 3.990 -1.841 6.881 0.331 1.494 -2.381
p-value 0.2400 0.1110 0.0000 0.1999 0.0001 0.0667 0.0000 0.7410 0.1355 0.0175
Table B 5.5: Significance of Differences between Country Means over Time After
Sales Service
- 102 -
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 4.514 3.887 -0.169 3.552 1.466 0.369 -2.929 1.432 4.131 5.805
p-value 0.0000 0.0001 0.8655 0.0004 0.1431 0.7123 0.0036 0.1529 0.0000 0.0000
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= -3.111 2.227 0.844 7.974 -1.334 1.749 -3.914 -0.383 1.960 3.614
p-value 0.0019 0.0265 0.3992 0.0000 0.1827 0.0814 0.0001 0.7018 0.0504 0.0003
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 0.840 6.909 -0.508 5.724 -1.116 1.419 1.228 -1.586 1.993 0.236
p-value 0.4013 0.0000 0.6116 0.0000 0.2648 0.1569 0.2204 0.1135 0.0466 0.8138
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 7.907 -11.066 4.746 11.677 6.400 2.231 1.708 -1.516 3.270 1.943
p-value 0.0000 0.0000 0.0000 0.0000 0.0000 0.0265 0.0884 0.1301 0.0011 0.0524
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= -0.992 4.758 2.955 2.287 4.310 4.234 -2.829 1.262 6.021 5.300
p-value 0.3216 0.0000 0.0032 0.0224 0.0000 0.0000 0.0049 0.2074 0.0000 0.0000
- 103 -
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 0.202 2.673 0.110 -2.090 -0.499 -0.221 -4.716 4.171 0.600 6.451
p-value 0.8403 0.0078 0.9123 0.0369 0.6180 0.8252 0.0000 0.0000 0.5484 0.0000
t Test: Significance of differences between country means over time (Q1 - Q4 2006)
Bots Egypt Ghana Kenya Maur Seych Tanz Ugan Zam Zim
Sp = 0.899 0.869 0.928 0.888 0.942 0.908 0.917 0.931 0.925 0.896
t= 0.755 4.545 -0.124 -1.897 1.048 1.336 1.133 -0.926 3.435 -0.860
p-value 0.4502 0.0000 0.9015 0.0581 0.2949 0.1828 0.2578 0.3550 0.0006 0.3899
- 104 -
APPENDIX C: DATA FREQUENCY TABLES
The data was summarised by tabulating the frequency (and proportion) of occurrence of
each preference permutation, as illustrated below.
Botswana Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 216 77 29 32 18 15 41 6 16 37 13 4 27 6 23
2 85 66 40 47 7 36 58 19 15 65 33 12 36 16 25
3 57 40 43 21 15 37 76 26 21 57 28 22 41 33 43
4 202 377 448 460 520 472 385 509 508 401 486 522 456 505 469
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.4 0.1 0.1 0.1 0 0 0.1 0 0 0.1 0 0 0 0 0
2 0.2 0.1 0.1 0.1 0 0.1 0.1 0 0 0.1 0.1 0 0.1 0 0
3 0.1 0.1 0.1 0 0 0.1 0.1 0 0 0.1 0.1 0 0.1 0.1 0.1
4 0.4 0.7 0.8 0.8 0.9 0.8 0.7 0.9 0.9 0.7 0.9 0.9 0.8 0.9 0.8
Egypt Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 99 86 19 23 10 24 11 0 5 25 1 1 4 2 6
2 53 55 23 26 19 38 19 7 11 29 12 4 13 1 6
3 27 31 25 20 22 44 29 8 20 38 16 4 15 10 7
4 137 144 249 247 265 210 257 301 280 224 287 307 284 303 297
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.3 0.1 0.1 0 0.1 0 0 0 0.1 0 0 0 0 0
2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0 0 0.1 0 0 0 0 0
3 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0 0 0 0
4 0.4 0.5 0.8 0.8 0.8 0.7 0.8 1 0.9 0.7 0.9 1 0.9 1 0.9
- 105 -
Ghana Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 128 52 61 76 17 34 35 9 7 33 13 5 8 11 7
2 65 64 39 68 16 54 45 25 27 23 14 19 16 9 12
3 47 41 43 40 26 35 52 23 28 26 49 24 22 19 21
4 256 339 353 312 437 373 364 439 434 414 420 448 450 457 456
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.1 0.1 0.2 0 0.1 0.1 0 0 0.1 0 0 0 0 0
2 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0.1 0 0 0 0 0 0
3 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0 0 0 0
4 0.5 0.7 0.7 0.6 0.9 0.8 0.7 0.9 0.9 0.8 0.8 0.9 0.9 0.9 0.9
Kenya Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 138 58 74 31 9 34 106 6 26 59 14 5 6 4 1
2 66 49 68 41 15 50 97 23 47 61 31 12 4 4 3
3 73 43 56 32 22 81 71 27 45 50 30 19 7 7 8
4 294 421 373 467 525 406 297 515 453 401 496 535 554 556 559
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.2 0.1 0.1 0.1 0 0.1 0.2 0 0 0.1 0 0 0 0 0
2 0.1 0.1 0.1 0.1 0 0.1 0.2 0 0.1 0.1 0.1 0 0 0 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0.1 0 0 0 0
4 0.5 0.7 0.7 0.8 0.9 0.7 0.5 0.9 0.8 0.7 0.9 0.9 1 1 1
Mauritius Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 64 56 15 47 25 32 33 13 10 31 23 19 2 3 7
2 45 48 26 30 27 44 22 16 13 35 25 18 11 8 12
3 31 27 19 34 28 38 25 21 15 38 40 25 13 10 16
4 240 249 320 269 300 266 300 330 342 276 292 318 354 359 345
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.2 0.1 0 0.1 0.1 0.1 0.1 0 0 0.1 0.1 0.1 0 0 0
2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0 0.1 0.1 0 0 0 0
3 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0 0 0
4 0.6 0.7 0.8 0.7 0.8 0.7 0.8 0.9 0.9 0.7 0.8 0.8 0.9 0.9 0.9
- 106 -
Seychelles Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 52 14 11 10 27 32 6 4 4 12 15 8 3 1 1
2 24 21 13 19 26 26 11 3 11 17 13 8 2 6 0
3 26 18 8 16 16 16 13 7 7 28 26 7 1 4 7
4 98 147 168 155 131 126 170 186 178 143 146 177 194 189 192
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.1 0.1 0.1 0.1 0.2 0 0 0 0.1 0.1 0 0 0 0
2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0 0 0 0
3 0.1 0.1 0 0.1 0.1 0.1 0.1 0 0 0.1 0.1 0 0 0 0
4 0.5 0.7 0.8 0.8 0.7 0.6 0.9 0.9 0.9 0.7 0.7 0.9 1 0.9 1
Tanzania Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 75 41 17 24 4 6 14 10 10 11 6 7 9 5 9
2 30 38 26 39 6 16 27 7 14 7 8 9 8 8 5
3 27 21 26 44 10 20 18 17 9 12 11 14 3 6 10
4 116 148 179 141 228 206 189 214 215 218 223 218 228 229 224
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.2 0.1 0.1 0 0 0.1 0 0 0 0 0 0 0 0
2 0.1 0.2 0.1 0.2 0 0.1 0.1 0 0.1 0 0 0 0 0 0
3 0.1 0.1 0.1 0.2 0 0.1 0.1 0.1 0 0 0 0.1 0 0 0
4 0.5 0.6 0.7 0.6 0.9 0.8 0.8 0.9 0.9 0.9 0.9 0.9 0.9 0.9 0.9
Uganda Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 86 24 48 29 3 11 35 4 13 38 9 4 6 5 15
2 49 39 31 30 17 26 35 11 14 29 7 6 13 12 11
3 45 32 34 24 9 23 30 13 18 27 15 10 11 15 24
4 150 235 217 247 301 270 230 302 285 236 299 310 300 298 280
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.1 0.1 0.1 0 0 0.1 0 0 0.1 0 0 0 0 0
2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0 0 0.1 0 0 0 0 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0 0 0 0 0.1
4 0.5 0.7 0.7 0.7 0.9 0.8 0.7 0.9 0.9 0.7 0.9 0.9 0.9 0.9 0.8
- 107 -
Zambia Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 166 52 45 27 13 14 49 9 29 30 9 16 10 1 18
2 37 44 58 62 23 39 70 16 35 37 17 21 9 9 11
3 40 39 34 29 16 34 44 19 26 50 37 33 28 19 40
4 245 353 351 370 436 401 325 444 398 371 425 418 441 459 419
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.3 0.1 0.1 0.1 0 0 0.1 0 0.1 0.1 0 0 0 0 0
2 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0 0 0 0 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0.1 0.1 0.1 0 0.1
4 0.5 0.7 0.7 0.8 0.9 0.8 0.7 0.9 0.8 0.8 0.9 0.9 0.9 0.9 0.9
Zimbabwe Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 112 56 51 16 20 35 81 6 20 59 12 7 6 5 8
2 65 62 49 27 22 28 87 10 46 54 14 3 10 5 12
3 67 38 51 28 15 38 57 14 33 61 23 14 20 20 15
4 250 338 343 423 437 393 269 464 395 320 445 470 458 464 459
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1 0.2 0.1 0.1 0 0 0.1 0.2 0 0 0.1 0 0 0 0 0
2 0.1 0.1 0.1 0.1 0 0.1 0.2 0 0.1 0.1 0 0 0 0 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0 0 0 0 0
4 0.5 0.7 0.7 0.9 0.9 0.8 0.5 0.9 0.8 0.6 0.9 1 0.9 0.9 0.9
- 108 -
Appendix C.2: Data Frequency Tables Q1 2006
The data was summarised by tabulating the frequency (and proportion) of occurrence of
each preference permutation, as illustrated below.
Botswana Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 102 51 15 27 6 16 33 3 13 28 6
2 49 53 24 23 10 42 48 6 9 29 7
3 35 28 22 32 10 34 37 6 23 42 31
4 114 168 239 218 274 208 182 285 255 201 256
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.3 0.2 0.1 0.1 0 0.1 0.1 0 0 0.1 0
2 0.2 0.2 0.1 0.1 0 0.1 0.2 0 0 0.1 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0.1
4 0.4 0.6 0.8 0.7 0.9 0.7 0.6 1 0.9 0.7 0.9
Egypt Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 51 32 6 6 3 9 3 22 5 13 2
2 21 35 8 16 9 24 4 21 0 13 1
3 19 26 10 12 10 15 7 24 5 18 6
4 61 59 128 118 130 104 138 85 142 108 143
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.3 0.2 0 0 0 0.1 0 0 0 0.1 0
2 0.1 0.2 0.1 0.1 0.1 0.2 0 0.1 0 0.1 0
3 0.1 0.2 0.1 0.1 0.1 0.1 0 0.2 0 0.1 0
4 0.4 0.4 0.8 0.8 0.9 0.7 0.9 0.6 0.9 0.7 0.9
- 109 -
Ghana Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 47 31 45 38 9 12 30 1 5 17 15
2 31 38 33 22 18 36 31 7 5 21 8
3 33 19 29 32 9 26 29 14 21 17 21
4 139 162 143 158 214 176 160 228 219 195 206
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.2 0.1 0.2 0.2 0 0 0.1 0 0 0.1 0.1
2 0.1 0.2 0.1 0.1 0.1 0.1 0.1 0 0 0.1 0
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0.1 0.1 0.1 0.1
4 0.6 0.6 0.6 0.6 0.9 0.7 0.6 0.9 0.9 0.8 0.8
Kenya Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 82 28 32 18 6 7 38 2 17 56 16
2 42 19 33 32 15 25 50 2 25 42 17
3 35 24 31 24 12 20 38 7 27 43 39
4 143 231 206 228 269 250 176 291 233 161 230
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.3 0.1 0.1 0.1 0 0 0.1 0 0.1 0.2 0.1
2 0.1 0.1 0.1 0.1 0 0.1 0.2 0 0.1 0.1 0.1
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0.1
4 0.5 0.8 0.7 0.8 0.9 0.8 0.6 1 0.8 0.5 0.8
Mauritius Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 44 33 12 12 12 30 21 2 5 16 13
2 37 21 12 19 20 26 21 7 9 16 12
3 24 15 7 21 12 27 23 7 4 36 24
4 95 131 169 148 156 117 135 184 182 132 151
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.2 0.2 0.1 0.1 0.1 0.2 0.1 0 0 0.1 0.1
2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0 0 0.1 0.1
3 0.1 0.1 0 0.1 0.1 0.1 0.1 0 0 0.2 0.1
4 0.5 0.7 0.8 0.7 0.8 0.6 0.7 0.9 0.9 0.7 0.8
- 110 -
Seychelles Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 25 13 0 6 15 3 1 1 0 7 4
2 5 3 5 13 12 17 6 2 1 4 7
3 5 5 1 7 6 8 7 3 3 17 13
4 40 54 69 49 42 47 61 69 71 47 51
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.3 0.2 0 0.1 0.2 0 0 0 0 0.1 0.1
2 0.1 0 0.1 0.2 0.2 0.2 0.1 0 0 0.1 0.1
3 0.1 0.1 0 0.1 0.1 0.1 0.1 0 0 0.2 0.2
4 0.5 0.7 0.9 0.7 0.6 0.6 0.8 0.9 0.9 0.6 0.7
Tanzania Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 33 45 13 10 9 7 8 7 7 10 3
2 16 19 11 10 13 19 20 10 15 13 6
3 12 6 4 10 8 31 13 8 20 31 9
4 91 82 124 122 122 95 111 127 110 98 134
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.2 0.3 0.1 0.1 0.1 0 0.1 0 0 0.1 0
2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0
3 0.1 0 0 0.1 0.1 0.2 0.1 0.1 0.1 0.2 0.1
4 0.6 0.5 0.8 0.8 0.8 0.6 0.7 0.8 0.7 0.6 0.9
Uganda Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 34 16 18 16 3 21 18 5 6 11 7
2 14 14 14 19 11 19 25 9 9 13 8
3 20 8 11 15 4 8 32 14 12 15 16
4 87 117 112 105 137 107 80 127 128 116 124
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.2 0.1 0.1 0.1 0 0.1 0.1 0 0 0.1 0
2 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 0.1
3 0.1 0.1 0.1 0.1 0 0.1 0.2 0.1 0.1 0.1 0.1
4 0.6 0.8 0.7 0.7 0.9 0.7 0.5 0.8 0.8 0.7 0.8
- 111 -
Zambia Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 94 30 24 21 5 12 26 6 6 23 3
2 34 41 34 28 11 18 29 8 14 18 15
3 31 36 22 23 12 26 35 12 12 28 13
4 91 143 170 178 222 194 160 224 218 181 219
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.4 0.1 0.1 0.1 0 0 0.1 0 0 0.1 0
2 0.1 0.2 0.1 0.1 0 0.1 0.1 0 0.1 0.1 0.1
3 0.1 0.1 0.1 0.1 0 0.1 0.1 0 0 0.1 0.1
4 0.4 0.6 0.7 0.7 0.9 0.8 0.6 0.9 0.9 0.7 0.9
Zimbabwe Attribute
Rank (Count) 1 2 3 4 5 6 7 8 9 10 11
1 81 37 17 14 3 8 55 3 13 16 4
2 41 48 34 24 10 22 31 6 10 13 12
3 35 14 20 29 9 25 43 9 12 31 24
4 94 152 180 184 229 196 122 233 216 191 211
Rank (Proportion) 1 2 3 4 5 6 7 8 9 10 11
1 0.3 0.1 0.1 0.1 0 0 0.2 0 0.1 0.1 0
2 0.2 0.2 0.1 0.1 0 0.1 0.1 0 0 0.1 0
3 0.1 0.1 0.1 0.1 0 0.1 0.2 0 0 0.1 0.1
4 0.4 0.6 0.7 0.7 0.9 0.8 0.5 0.9 0.9 0.8 0.8
- 112 -